{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9004832647216753,
  "eval_steps": 1677,
  "global_step": 15093,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 1.9403376886820598e-05,
      "loss": 3.0319,
      "step": 500
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.8806753773641194e-05,
      "loss": 2.8898,
      "step": 1000
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.8210130660461786e-05,
      "loss": 2.8471,
      "step": 1500
    },
    {
      "epoch": 0.1,
      "eval_f1": 0.09119675351283209,
      "eval_loss": 2.799565315246582,
      "eval_runtime": 565.0093,
      "eval_samples_per_second": 210.945,
      "eval_steps_per_second": 3.297,
      "step": 1677
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.7613507547282383e-05,
      "loss": 2.826,
      "step": 2000
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.701688443410298e-05,
      "loss": 2.8064,
      "step": 2500
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6420261320923575e-05,
      "loss": 2.7912,
      "step": 3000
    },
    {
      "epoch": 0.2,
      "eval_f1": 0.09886291342405676,
      "eval_loss": 2.7552475929260254,
      "eval_runtime": 565.8136,
      "eval_samples_per_second": 210.645,
      "eval_steps_per_second": 3.293,
      "step": 3354
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.582363820774417e-05,
      "loss": 2.769,
      "step": 3500
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.5227015094564765e-05,
      "loss": 2.7626,
      "step": 4000
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.463039198138536e-05,
      "loss": 2.7493,
      "step": 4500
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.4033768868205956e-05,
      "loss": 2.7485,
      "step": 5000
    },
    {
      "epoch": 0.3,
      "eval_f1": 0.10934945582171454,
      "eval_loss": 2.701012134552002,
      "eval_runtime": 563.4598,
      "eval_samples_per_second": 211.525,
      "eval_steps_per_second": 3.306,
      "step": 5031
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.343714575502655e-05,
      "loss": 2.7355,
      "step": 5500
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.2840522641847146e-05,
      "loss": 2.7254,
      "step": 6000
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.2243899528667742e-05,
      "loss": 2.7163,
      "step": 6500
    },
    {
      "epoch": 0.4,
      "eval_f1": 0.11683918795621939,
      "eval_loss": 2.6925740242004395,
      "eval_runtime": 561.7517,
      "eval_samples_per_second": 212.168,
      "eval_steps_per_second": 3.316,
      "step": 6708
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.1647276415488338e-05,
      "loss": 2.7101,
      "step": 7000
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.1050653302308931e-05,
      "loss": 2.7063,
      "step": 7500
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.0454030189129527e-05,
      "loss": 2.7037,
      "step": 8000
    },
    {
      "epoch": 0.5,
      "eval_f1": 0.1264069487053307,
      "eval_loss": 2.657057762145996,
      "eval_runtime": 557.0113,
      "eval_samples_per_second": 213.974,
      "eval_steps_per_second": 3.345,
      "step": 8385
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.857407075950123e-06,
      "loss": 2.6922,
      "step": 8500
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.260783962770719e-06,
      "loss": 2.6857,
      "step": 9000
    },
    {
      "epoch": 0.57,
      "learning_rate": 8.664160849591315e-06,
      "loss": 2.6796,
      "step": 9500
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.06753773641191e-06,
      "loss": 2.6686,
      "step": 10000
    },
    {
      "epoch": 0.6,
      "eval_f1": 0.1275429879642049,
      "eval_loss": 2.6493475437164307,
      "eval_runtime": 556.9366,
      "eval_samples_per_second": 214.003,
      "eval_steps_per_second": 3.345,
      "step": 10062
    },
    {
      "epoch": 0.63,
      "learning_rate": 7.470914623232505e-06,
      "loss": 2.6798,
      "step": 10500
    },
    {
      "epoch": 0.66,
      "learning_rate": 6.874291510053101e-06,
      "loss": 2.6528,
      "step": 11000
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.277668396873695e-06,
      "loss": 2.6469,
      "step": 11500
    },
    {
      "epoch": 0.7,
      "eval_f1": 0.1396037444667752,
      "eval_loss": 2.620657444000244,
      "eval_runtime": 557.3053,
      "eval_samples_per_second": 213.861,
      "eval_steps_per_second": 3.343,
      "step": 11739
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.681045283694291e-06,
      "loss": 2.6569,
      "step": 12000
    },
    {
      "epoch": 0.75,
      "learning_rate": 5.0844221705148865e-06,
      "loss": 2.6472,
      "step": 12500
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.487799057335481e-06,
      "loss": 2.6507,
      "step": 13000
    },
    {
      "epoch": 0.8,
      "eval_f1": 0.1391354358365679,
      "eval_loss": 2.6126439571380615,
      "eval_runtime": 556.9611,
      "eval_samples_per_second": 213.993,
      "eval_steps_per_second": 3.345,
      "step": 13416
    },
    {
      "epoch": 0.81,
      "learning_rate": 3.891175944156077e-06,
      "loss": 2.6297,
      "step": 13500
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.294552830976672e-06,
      "loss": 2.6364,
      "step": 14000
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.6979297177972674e-06,
      "loss": 2.6287,
      "step": 14500
    },
    {
      "epoch": 0.89,
      "learning_rate": 2.101306604617863e-06,
      "loss": 2.6269,
      "step": 15000
    },
    {
      "epoch": 0.9,
      "eval_f1": 0.14388305284850214,
      "eval_loss": 2.6065549850463867,
      "eval_runtime": 556.552,
      "eval_samples_per_second": 214.151,
      "eval_steps_per_second": 3.347,
      "step": 15093
    }
  ],
  "logging_steps": 500,
  "max_steps": 16761,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1677,
  "total_flos": 4.1849662176682214e+17,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}